THE DETECTION OF PSYCHIATRIC MORBIDITY AND ITS EFFECTS ON OUTCOME IN ACUTE ELDERLY MEDICAL ADMISSIONS

This study examines the point prevalence of psychiatric morbidity in patients admitted to acute geriatric care, the performance of screening questionnaires in detecting psychiatric morbidity and the relationship between psychiatric morbidity and outcome (in terms of length of hospital admission and mortality) after controlling for severity of physical illness. In a sample of 119 consecutive admissions, 65 (61%) of the 106 patients assessed had dementia, while nine (10%) of the 88 assessed had a diagnosis of depression and a further 35 (40%) significant depressive symptoms, and 22 (20%) of the 106 assessed had delirium. Screening with the Mini-Mental State Examination had 81% sensitivity and 83% specificity for dementia. The Geriatric Depression Scale (GDS) had 74% sensitivity and 72% specificity for depressive symptoms; both the GDS and General Health Questionnaire had 100% sensitivity for depressive illness. Mortality was significantly higher in the delirious (62%) than in the non-delirious (14%, p < 0.001) and in the severely demented (65%) than the mildly demented (29%) and non-demented (15%), p < 0.001) subjects. For delirium this effect was found to be independent of the severity of physical illness. Patients with severe dementia had significantly longer hospital admissions (p < 0.02), No relationship was found between depression and outcome.